So, as I understand the general Gaf opinion is based on the Aqua/Cream leaks over the months. It´s a little hard to find their posts on those threads, as it is restricted to registered members.

It would probaly easier find their numbers here... I will check.

Any way, my point is that there are contradition from each different leaker, eventhough they make a effort to "consolidate" their numbers. That is when I start to suspect that they only do a more detailed inference with past data. That, I MY OPINION, would explain why the leaks took some time to be... leaked.

Those french numbers could be right or more right than the general opinion I´ve been reading there.

Beautiful job Boutrous with the thread.

Kresnik, the reason for the bad acronyms and misspellings is to try to dodge NPD ninjas from simply searching for a term like 'NPD Thread + Hyrule' and locating the leaker that easily. It's probably a bit futile masking things like that, I imagine that with a $$$$ outfit like they run that they have multiple people reading the entire thread, and probably dossiers on many Gaffers (and maybe even a few of us here, lol).

If they can tie a leaker to a real name, and find out who they work for, where they got the data, then they can try to contact the contract-holder at the company where the data was leaked out of and get some retribution going. IIRC a few months ago some web site leaked the entire thing in clear text, and it got kicked to the moon because of it (they knew where it came from somehow, probably by altering small details in the data so as to to know preceisely who got THAT dataset). What do I mean by that? Maybe data that you release to (insert corporation here) has 538,650 listed as sales for X, and data that you release to (insert corporation here) has 538,645 listed for the same item. Further, you can alter punctuation, spacing, and other things slightly in automated ways to 'riskier' or suspected customers so as to catch anyone dumb enough to simply copy/paste it somewhere.

That's why you see a lot of pie charts, abbreviated data like > 800 and < 750, etc. NPD Ninjas would love to have creamsugar's head on a pike, and any other leakers they can get.

Beautiful job Boutrous with the thread.

At that time it was a commun statemant to assume the gap was between 600k and 750k. Last year gap was only 100k. We also knew PS4 sold almost 2x the XOne in January (150k?), February was close (100k?) and March closer(50k?), them for April, May and June was a given there that PS4 beated by close to January numbers and ratio.

http://www.forbes.com/sites/erikkain/2014/07/18/npd-hardware-sales-numbers-leak-for-june-2014/

Total year to date:

• PS4 – 1,577,000
• XB1 – 1,101,000
• 3DS – 768,000
• 360 – 464,000
• Wii U – 448,000
• PS3 – 337,000
• Vita – 128,000
  
  

Edit: The best I could find about the consolidate opinion on Neogaf.

For May: 

– Wolfenstein: The New Order platform breakdown: 41% PlayStation 4, 38% Xbox One, 13% Xbox 360, 8% PlayStation 3
– Minecraft: PlayStation 3 Edition sales: 89,000 units
– 370k < ∞ < 380k, bundle ~ 6.6%
- Wii U + Xbox family < PS4
- 200k > PS4 > 100k > XBO > Wii U > 360 > 50k > PS3 > Wii

to mornelithe

He's not understanding. There are multiple gaffers with explicit access to NPD numbers, but they can't just put them out there in plain text, they have to obfuscate a bit. That's why you see graphs, yes/no answers, etc.

The final consensus is always solid though, and you would get banned to the moon for lying about the numbers. But instead of "124k" for an item, you might get ">120 and <130" Etc.

Basically you have a lock within a scant % on everything major on a reliable basis. Some months are extra tough due to the most connected leakers being careful or taking their sweet time with it.